Optimal Operation of Distribution Networks with
Synchronous Generators via Transient Stability
Constrained Optimal Power Flow
Kamile Fuchs, Roman Kuiava and Thelma S. P. Fernandes
Department of Electrical Engineering
Federal University of Parana (UFPR)
Curitiba, Brazil
Email: kamilef@yahoo.com.br, kuiava@eletrica.ufpr.br, thelma@eletrica.ufpr.br
Abstract— A common approach to deal with transient stability
constraints in optimization problems is the combination of a
numerical discretization method with interior point methods
(IPMs) for solving large-scale nonlinear programming (NLP).
The numerical discretization is adopted to convert the differential
equations that describe the electromechanical transients in power
systems in a set of algebraic equations to be included into a
conventional optimal power flow (OPF) problem. The resulting
transient stability constrained optimal power flow (TSC-OPF)
problem, however, suffers with the curse of dimensionality, high
computational time and memory consumption to solve it, even for
small systems. To relieve this computational burden, this paper
proposes an algorithm that aims at solving a TSC-OPF problem
via primal-dual IPM with only a few time steps of numerical
discretization in post-fault period, which is enough to ensure the
rotor angle first swing stability. Once the TSC-OPF problem is
solved, a numerical discretization method is applied to calculate
the system trajectory from the first rotor angle peak in post-fault
period. An important contribution of the proposed algorithm is
to provide a proper accuracy on the computation of constant
admittance loads. Numerical results are obtained for a 9-bus
distribution network with the purpose of providing a proper
sizing of DG units based on synchronous generators and an
optimal operation to the network.
I. I NTRODUCTION
The connection of distributed generation (DG) units in
sub-transmission and distribution networks poses a challenge
for the system operation and management in medium and
low voltages. With the new environment provided by the
inclusion of small generators close to loads, many researches
have been focused on certain issues that were not being
discussed previously when the distribution networks were
basically passive and radial circuits. Some potential impacts
of DG units on the distribution operation can be observed, for
example, as changes on the power quality and the presence of
low-frequency electromechanical oscillations, as discussed in
[1]–[5].
Furthermore, an inappropriate siting and sizing of DGs units
can cause either under-voltage or over-voltage violations in the
distribution feeders [6], as well as, an inadequate transient
This work was supported by Institutos LACTEC.
response performance of these generators with respect to
severe perturbations. In this context, the optimal power flow
(OPF) has been an important tool for studies involving the
proper locations and sizing of DG units and optimal operation
of distribution networks. The set of constraints considered in
the OPF is generally determined by only physical and static
operational limits. However, in recent years, for reasons of
protection against occurrences of severe disturbances, OPF
problems involving dynamic security, such as transient stabil-
ity constraints, have received considerable research attention,
as verified in [7]–[10].
The strategy of incorporating transient stability constraints
within an OPF problem poses the challenge of joining time-
simulation and optimization. The mathematical modeling of
a transient stability constrained optimal power flow (TSC-
OPF) becomes very complex due to highly non-linear nature of
the differential equations that describes the electromechanical
transient behavior of synchronous generators. Another limita-
tion is the possibility of incurring a computational convergence
problem since the region of feasible solutions becomes very
restricted.
Based on these considerations, this paper proposes an
algorithm that aims at solving a TSC-OPF problem with a
minimum number of time steps of numerical discretization
in post-fault period, which is enough to ensure the rotor
angle first swing stability (FSS). In this paper, the Primal-
Dual IPM is incorporated to the proposed algorithm to solve
the TSC-OPF. One important contribution of the proposed
algorithm is to provide a proper accuracy on the computation
of constant admittance loads. The computation of these loads
admittances is relevant in the process of solving these opti-
mization problems, once that steady-state bus voltages (either
magnitude or angle) are dependent variables to be determined
by the solution of the TSC-OPF problem, which means they
are not previously known for the computation of the system
loads admittance. Once the TSC-OPF problem is solved, a
numerical discretization method is applied to calculate the
system trajectory from the first rotor angle peak in post-fault
period.
978-1-4673-8040-9/15/$31.00 ©2015 IEEE